High-Flow Nasal Cannula and Standard Oxygen in Acute Hypoxemic Respiratory Failure Due to COVID-19

BACKGROUND: The evidence regarding benefits of high-flow nasal cannula (HFNC) in patients with COVID-19 is controversial. The aim of this study was to evaluate the impact of HFNC in comparison with standard oxygen therapy on the frequency of endotracheal intubation at 28 d in subjects with acute hypoxemic respiratory failure (AHRF) secondary to SARS-CoV-2 infection.

METHODS: A retrospective, age- and sex-matched–paired, cohort study was conducted in subjects with moderate-to-severe AHRF. Intervention group was treated with HFNC, and control group was treated with standard oxygen therapy. Baseline characteristics and clinical evolution were analyzed. Mantel-Haenszel test was used for categorical variables. Paired samples Wilcoxon test was used for quantitative variables. Multivariate analysis was performed using conditional multiple logistic regression.

RESULTS: Eighty-four subjects were included. The median time from admission to progression of oxygen therapy to F_IO2 ≥ 0.5 or HFNC was 1 (interquartile range [IQR] 0–3) d. P_aO2/F_IO2 at the time of oxygen therapy progression showed a median of 150.5 (IQR 100.0–170.0) for the entire sample and was lower in HFNC group compared with control group (median 135 [IQR 96–162] vs median 158 [IQR 132–174], respectively, P = .02). Endotracheal intubation at 28 d was observed in 54.8% HFNC and 73.8% standard oxygen (unadjusted odds ratio 0.38 [95% CI 0.13–1.07], P = .069). In the multivariate analysis, presence of dyspnea at hospital admission, Sequential Organ Failure Assessment score, and P_aO2/F_IO2 at time of progression of oxygen therapy to F_IO2 ≥ 0.5 was identified as confounding factors for the association between the intervention group and the outcome. Use of HFNC was not an independent predictor of endotracheal intubation frequency after adjusting confounders (odds ratio 0.26 [95% CI 0.04–1.51], P = .13).

CONCLUSIONS: In this study, HFNC therapy in subjects with AHRF secondary to COVID-19 was not an independent predictor of endotracheal intubation, compared with standard oxygen therapy, after adjusting for confounders.